Vignettes revisited

[antagomir]

Now as the PRs have been merged and simulator functions output TreeSE objects, we should update the vignettes so that they use the TreeSE outputs.

This is potentially complex task, and not necessarily everything has to be redone because the other option in some cases is to either remove/simplify examples, or pick the needed data items (abundance matrices, sample metadata) separately from the TreeSE then just using whatever worked earlier.

But we need to work through this case by case and PRs should be limited in scope. Otherwise managing them can get out of hands.

Can we have PRs separate PR for each vignette. One by one. Perhaps even one chunk at a time before going to the next ones, if needed. We should consider carefully whether the vignettes should contain the full case studies because they are computationally heavy and cause problems in package builds. Another option is to move some complex workflows into additional data files in inst/extras/. For now, make sure that the vignette examples use the TreeSE objects as output from simulator functions. Fix straightforward cases and for more complex cases let us discuss in more detail.

[antagomir]

Also pay attention to simplifying and speeding up examples. The long running times are problematic for automated builds and tests.

[antagomir]

If we need to include full case studies that are slow to compute, we need to device an alternative way for those, to build locally before release. We could check if there is a handy way of doing that with pkgdown websites.

[antagomir]

Solved now by separating the original case studies from the later TreeSE-based material-